CN106934502A - A kind of movement in stock and share rate Forecasting Methodology - Google Patents
A kind of movement in stock and share rate Forecasting Methodology Download PDFInfo
- Publication number
- CN106934502A CN106934502A CN201710173259.XA CN201710173259A CN106934502A CN 106934502 A CN106934502 A CN 106934502A CN 201710173259 A CN201710173259 A CN 201710173259A CN 106934502 A CN106934502 A CN 106934502A
- Authority
- CN
- China
- Prior art keywords
- stock
- factor
- data
- assumptions
- risks
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The present invention relates to technical field of information processing, and in particular to a kind of movement in stock and share rate Forecasting Methodology.Comprise the following steps：Step one, from database read initial data, initial data include the market data and financial statement data relevant with stock；Step 2, set up stability bandwidth forecast model；Step 3, the model calculating personal share volatitle revenue dynamic rate using foundation in step 2.The present invention makes prediction according to the predictive risk following to investment combination of model, so that risk management before helping investor preferably to do good.The present invention can be according to different Factorizations, can not only predict and quantify risk, the stability of risk profile is improved, and investment risk can be decomposed separate sources, selected to undertake which risk so as to make investor more targeted and which risk evaded.
Description
Technical field
The present invention relates to technical field of information processing, and in particular to a kind of movement in stock and share rate Forecasting Methodology.
Background technology
Stability bandwidth refers to the intensity of variation of target assets investment return rate, have actual volatility and Historic Volatility point.
Actual volatility is also referred to as following stability bandwidth, and it refers to the measurement to equity investment return rate degree of fluctuation, due to investment repayment
Rate is a random process, and actual volatility is forever a unknown number.In other words, actual volatility is accurately to count in advance
Calculate, people can only obtain its estimate by various methods.Historic Volatility referred to rate of return on investment in one section of past
Interior shown stability bandwidth, it by target property market price the past period historical data (the i.e. time sequence of St
Row data) reflection.In other words, corresponding stability bandwidth data, Ran Houyun can be calculated according to the time series data of { St }
The standard deviation of return rate is estimated with Statistical Inference, so as to obtain the estimate of Historic Volatility.Obviously, if actual fluctuation
Rate is a constant, and it does not change with the passage of time, then Historic Volatility be possible to be one of actual volatility very
Good is approximate.
Current China's stock type fund, mainly the solution in risk management is risk management afterwards.Risk afterwards
Management is the operating mechanism of bin level management, and its core is an information processing mechanism, the industry of monitoring fund or investment combination
Achievement, and parameter is stopped loss according to the risk for setting in advance determine to the treatment of the position in storehouse of investment combination.
Publication No. CN105989535A, publication date is：2016.10.05 Chinese invention patent, discloses a B shareB
Stability bandwidth Forecasting Methodology and system, methods described comprise the following steps：Exchange quotation data are built when being closed daily according to each stock
Vertical Stock Risk Controlling model database；Plate weight shared by the corresponding stock of the stock code of reception input, stock code, if
Put the beginning and ending time of stock price tendency；Exchange quotation when being closed previously according to proxima luce (prox. luc) is read from risk control model database
The risk control parameter value that data are calculated；Plate weight, stock valency according to shared by stock code, stock code correspondence stock
The beginning and ending time of lattice tendency and/or Stock Risk control parameter value calculate stock and stock portfolio in stock price tendency
Beginning and ending time in stability bandwidth.
The movement in stock and share rate Forecasting Methodology of the disclosure of the invention is also based on the operating mechanism of risk management afterwards, there is modeling
Process is complicated, the problems such as result of calculation is inaccurate.
The content of the invention
It is an object of the present invention to be directed to problems of the prior art, propose that one kind is applied to risk management in advance
Movement in stock and share rate Forecasting Methodology, it is complicated to solve calculating process in the prior art, the inaccurate problem of result of calculation.
To achieve these goals, the technical solution adopted by the present invention is：
A kind of movement in stock and share rate Forecasting Methodology, comprises the following steps：
Step one, initial data is read from database, initial data includes that market data relevant with stock and finance are reported
Table data；
Step 2, set up stability bandwidth forecast model；
Step 3, the model calculating personal share volatitle revenue dynamic rate using foundation in step 2.
The computational methods of model are as follows in the step 2：
Assuming that the common k risks and assumptions of the multiple-factor risk model of earning rate.The then earning rate r of stock iiFor：
ri=bi1f1+bi2f2+bi3f3+....+bikfk+∈i (1)
Wherein bijIt is exposures of the stock i to risks and assumptions j；fjIt is the earning rate of risks and assumptions j, ∈iIt is the personal share of stock i
Income, i.e., expose that unrelated a part of income of brought income to risks and assumptions, different from the income of stock i with the stock
Rate ri；
Based on (1), the volatitle revenue dynamic rate of the stock is：
Wherein biIt is the k × 1 risks and assumptions exposure vector of stock i, V is k × k risks and assumptions yield variance-covariance square
Battle array, σεiThe stability bandwidth of personal share income, i.e. personal share risk.
The exposure computational methods of the risks and assumptions are：Take Z values to calculate to be standardized, in each day of trade knot
After beam initial data updates, every primitive factor exposure of stock is collected and extracted, it is assumed that stock i exposes as B to factor k'sik,
So standardization of the stock i to factor k is exposed as：
bik=(Bik-mean(Bk))/std(Bk) (3)
Wherein BikBe stock i to the standardization of factor k exposure, B be in sample all stocks to factor k it is exposed to
Amount.
The earning rate calculation of the risks and assumptions is：With the multiple power compound interest income of every stock of 22 day of trade of past
Rate is exposed as independent variable as dependent variable with the factor acquired before 22 day of trade, then does linear returning based on these data
Return.
Risks and assumptions yield variance-covariance matrix the calculation procedure is：
After step one, the closing quotation of each day of trade, newest initial data is obtained；
Step 2, calculating factor exposure, then calculate same day factor earning rate；
Step 3, with the factor earning rate time series data in past 36 months come calculation risk factor variance-covariance
Matrix.
The method for building up of the stability bandwidth forecast model is comprised the following steps：
Step one, initial data is read from database, initial data includes that market data relevant with stock and finance are reported
Table data；
Step 2, according to initial data build risks and assumptions；
Step 3, the factor that will be built in step 2 are stored into factor data storehouse；
Step 4, with step 3 determine risks and assumptions set up stability bandwidth forecast model.
The risks and assumptions include：Reverse the factor, factor of momentum, the market value factor, flowing sex factor, the debt ratio factor, city
The net rate factor and the industry factor；The reverse factor includes that stock goes over to weigh compound interest overall yield again in 1 month；The factor of momentum
Including stock go over 12 months again power compound interest overall yield subtract over one month again power compound interest overall yield difference；The city
The value factor includes negotiable stock price is multiplied by the value obtained after the product extraction of square root of negotiable stock quantity；The mobility because
Attached bag includes the transaction turnover rate that stock went over for 30 day of trade, and turnover rate is quotient of the trading volume divided by negotiable total amount；It is described
The debt ratio factor includes quotient of the previous year enterprise total liability divided by previous year enterprise shareholding equity；The HSBC factor bag
Include quotient of the previous year enterprise total market capitalisation divided by previous year enterprise shareholding equity；The industry factor includes being based on each plate
The historical data in past 36 months does Multivariate Time Series recurrence to every stock, and the plate includes can active plate, material
Plate, industrial plate, durable goods plate, optional commodity plate, medical treatment ＆ health plate, financial plate, information technology plate and
Public utilities plate.
The market data include the daily price data of stock and trading volume data.
The financial statement data includes the profit and loss statement data of each financial season of stock and wealth year, assets debt tables data
And cash flow statement data.
By adopting the above-described technical solution, the beneficial effects of the invention are as follows：
The present invention makes prediction according to the predictive risk following to investment combination of model, so as to help investor more preferable
Do good before risk management.It is different from the current technical problem for mainly focusing on risk management afterwards.In the present invention
Risk can not only be predicted and quantified to model according to different Factorizations, improve the stability of risk profile, and can be by
Investment risk decomposes separate sources, selects to undertake which risk so as to make investor more targeted and which wind evaded
Danger.
Specific embodiment
Inventing type is described in detail below.
Embodiment 1
As a kind of preferred embodiment of the invention, present embodiment discloses a kind of movement in stock and share rate Forecasting Methodology, including
Following steps：
Step one, initial data is read from database, initial data includes that market data relevant with stock and finance are reported
Table data；
Step 2, set up stability bandwidth forecast model；
Step 3, the model calculating personal share volatitle revenue dynamic rate using foundation in step 2.
The computational methods of model are as follows in the step 2：
Assuming that the common k risks and assumptions of the multiple-factor risk model of earning rate.The then earning rate r of stock iiFor：
ri=bi1f1+bi2f2+bi3f3+....+bikfk+∈i (1)
Wherein bijIt is exposures of the stock i to risks and assumptions j；fjIt is the earning rate of risks and assumptions j, ∈iIt is the personal share of stock i
Income, i.e., expose that unrelated a part of income of brought income to risks and assumptions, different from the income of stock i with the stock
Rate ri；
Based on (1), the volatitle revenue dynamic rate of the stock is：
Wherein biIt is the k × 1 risks and assumptions exposure vector of stock i, V is k × k risks and assumptions yield variance-covariance square
Battle array, σεiThe stability bandwidth of personal share income, i.e. personal share risk.
The exposure computational methods of the risks and assumptions are：Take Z values to calculate to be standardized, in each day of trade knot
After beam initial data updates, every primitive factor exposure of stock is collected and extracted, it is assumed that stock i exposes as B to factor k'sik,
So standardization of the stock i to factor k is exposed as：
bik=(Bik-mean(Bk))/std(Bk) (3)
Wherein BikBe stock i to the standardization of factor k exposure, B be in sample all stocks to factor k it is exposed to
Amount.
The earning rate calculation of the risks and assumptions is：With the multiple power compound interest income of every stock of 22 day of trade of past
Rate is exposed as independent variable as dependent variable with the factor acquired before 22 day of trade, then does linear returning based on these data
Return.
Risks and assumptions yield variance-covariance matrix the calculation procedure is：
After step one, the closing quotation of each day of trade, newest initial data is obtained；
Step 2, calculating factor exposure, then calculate same day factor earning rate；
Step 3, with the factor earning rate time series data in past 36 months come calculation risk factor variance-covariance
Matrix.
The factor exposure that factor variance-covariance matrix combines every stock can calculate every stock according to formula (2)
The system risk prediction of ticket.Regression residuals in linear regression are personal share factors, the every stock returned with past 36 months
The standard deviation of residual error weigh every personal share risk profile of stock.
The method for building up of the stability bandwidth forecast model is comprised the following steps：
Step one, initial data is read from database, initial data includes that market data relevant with stock and finance are reported
Table data；
Step 2, according to initial data build risks and assumptions；
Step 3, the factor that will be built in step 2 are stored into factor data storehouse；
Step 4, with step 3 determine risks and assumptions set up stability bandwidth forecast model；
The risks and assumptions include：Reverse the factor, factor of momentum, the market value factor, flowing sex factor, the debt ratio factor, city
The net rate factor and the industry factor；The reverse factor includes that stock goes over to weigh compound interest overall yield again in 1 month；The factor of momentum
Including stock go over 12 months again power compound interest overall yield subtract over one month again power compound interest overall yield difference；The city
The value factor includes negotiable stock price is multiplied by the value obtained after the product extraction of square root of negotiable stock quantity；The mobility because
Attached bag includes the transaction turnover rate that stock went over for 30 day of trade, and turnover rate is quotient of the trading volume divided by negotiable total amount；It is described
The debt ratio factor includes quotient of the previous year enterprise total liability divided by previous year enterprise shareholding equity；The HSBC factor bag
Include quotient of the previous year enterprise total market capitalisation divided by previous year enterprise shareholding equity；The industry factor includes being based on each plate
The historical data in past 36 months does Multivariate Time Series recurrence to every stock, and the plate includes can active plate, material
Plate, industrial plate, durable goods plate, optional commodity plate, medical treatment ＆ health plate, financial plate, information technology plate and
Public utilities plate.
For the calculating of the industry factor, using Fama-Macbeth methods, the historical data based on past 36 months is to every
Stock does Multivariate Time Series recurrence.And in trade division, using Wind first-level class modes, but due to telecommunication path
Plate is very few in some time interval stock quantity, so 9 plates are only have chosen, and the independent variable of time series also accordingly sets
It is set to 9.Advantage of this is that, we are objective to consider every stock may have the possibility of risk exposure to multiple industries
Property so that our analyses of model are more comprehensive.Then model can be according to such stock for drawing of analyzing to the sudden and violent of plate
Dew, then gross section regression is done together with the factor exposure of other risks and assumptions, obtain being mapped in the exposure of the industry-by-industry factor
On regression coefficient, and be based ultimately upon these regression coefficients and calculate the plate factor current yield.
The market data include the daily price data of stock and trading volume data.
The financial statement data includes the profit and loss statement data of each financial season of stock and wealth year, assets debt tables data
And cash flow statement data.
Realize that a kind of movement in stock and share rate forecasting system of the invention includes：
Stock quote information receiver module, the market data for receiving and updating each day of trade all A-shares automatically；
Stock fundamental data receiver module, the fundamental data for periodically receiving and updating all A-shares；
Database integration cleaning modul, for being docked and being matched stock quote information and stock basic side information,
And clear up and correct the mistake or deviation in raw data base；
Stability bandwidth forecast model analyze and update module, for setting up stability bandwidth forecast model, data are analyzed and
Computing show that risk data and data update.
The forecasting system also includes：
Risk model database module, for receiving the risk letter obtained by the analysis of stability bandwidth forecast model and update module
Breath, including risks and assumptions exposure, risks and assumptions earning rate, and risks and assumptions earning rate variance-covariance matrix, and it is stored into this
The risk information database of system.
The database is automatically updated in each day of trade.
User data receiver module, for receiving customer investment combined information；
The system supports that user is manually entered and also supports user input Excel.Customer investment combined information mainly includes using
Every ratio for holding ratio and remaining cash of stock in the investment combination of family.
Customer investment constitution's risk analysis module, the data according to user data receiver module are in conjunction with risk model data
The data that library module day updates just can immediately analyze the risk and stability bandwidth of customer investment combination.
User can be traded analysis in the module, change storehouse analysis and other analyses and prediction related to risk.
The market data include pricing information, except power ex dividend information and turnover information.
The fundamental data includes the profit and loss statement of financial season and wealth year, assets debt tables and cash flow statement；Also wrap
Prediction data of each stock trader analyst to net profit and the net earnings per share profit of stock is included, while also including every stock industry point
Category information.
The database integration and cleaning modul include two submodules：Data Integration submodule and data scrubbing submodule
Block.Data Integration submodule is mainly stock quote information and stock basic side information is docked and matched, so that being
System database accomplishes that day updates；The major function of data scrubbing submodule be clear up and correct mistake in raw data base or
Deviation, such as before peep deviation etc., so that it is guaranteed that the quality of data is accurate.
The operation that the stability bandwidth forecast model analysis and update module are followed the steps below：
Step one, initial data is extracted from the output of database integration cleaning modul；
Step 2, sets up stability bandwidth forecast model,
Assuming that the common k risks and assumptions of the multiple-factor risk model of earning rate.The then earning rate r of stock iiFor：
ri=bi1f1+bi2f2+bi3f3+....+bikfk+∈i (1)
Wherein bijIt is exposures of the stock i to risks and assumptions j；fjIt is the earning rate of risks and assumptions j, ∈iIt is the personal share of stock i
Income, i.e., expose that unrelated a part of income of brought income to risks and assumptions, different from the income of stock i with the stock
Rate ri；
Based on (1), the volatitle revenue dynamic rate of the stock is：
Wherein biIt is the k × 1 risks and assumptions exposure vector of stock i, V is k × k risks and assumptions income covariance matrixes, σεi
The stability bandwidth of personal share income, i.e. personal share risk；
Step 3, the exposed calculating of risks and assumptions,
After each transaction end of day initial data updates, every primitive factor exposure of stock is collected and extracted, it is assumed that
Stock i exposes as B to factor k'sik, then standardization of the stock i to factor k is exposed as：
bik=(Bik-mean(Bk))/std(Bk) (3)
Wherein BikBe stock i to the standardization of factor k exposure, B be in sample all stocks to factor k it is exposed to
Amount.
Step 4, the calculating of risks and assumptions earning rate,
With the multiple power compound interest earning rate of every stock of 22 day of trade of past as dependent variable, with being obtained before 22 day of trade
The factor for taking is exposed as independent variable, then does linear regression based on these data；
Step 5, risks and assumptions yield variance-covariance matrix is calculated,
After the closing quotation of each day of trade, newest initial data is obtained；Factor exposure is calculated, the same day factor is then calculated and is received
Beneficial rate；With the factor earning rate time series data in past 36 months come calculation risk factor variance-covariance matrix.
The initial data includes market data, stock trade classification information data, each financial season of stock and wealth year
Profit and loss statement data, assets debt tables data and cash flow statement data.
The risks and assumptions include：Reverse the factor, factor of momentum, the market value factor, flowing sex factor, the debt ratio factor, city
The net rate factor and the industry factor；The reverse factor includes that stock goes over to weigh compound interest overall yield again in 1 month；The factor of momentum
Including stock go over 12 months again power compound interest overall yield subtract over one month again power compound interest overall yield difference；The city
The value factor includes negotiable stock price is multiplied by the value obtained after the product extraction of square root of negotiable stock quantity；The mobility because
Attached bag includes the transaction turnover rate that stock went over for 30 day of trade, and turnover rate is quotient of the trading volume divided by negotiable total amount；It is described
The debt ratio factor includes quotient of the previous year enterprise total liability divided by previous year enterprise shareholding equity；The HSBC factor bag
Include quotient of the previous year enterprise total market capitalisation divided by previous year enterprise shareholding equity；The industry factor includes being based on each plate
The historical data in past 36 months does Multivariate Time Series recurrence to every stock, and the plate includes can active plate, material
Plate, industrial plate, durable goods plate, optional commodity plate, medical treatment ＆ health plate, financial plate, information technology plate and
Public utilities plate.
For the calculating of the industry factor, using Fama-Macbeth methods, the historical data based on past 36 months is to every
Stock does Multivariate Time Series recurrence.And in trade division, using Wind first-level class modes, but due to telecommunication path
Plate is very few in some time interval stock quantity, so 9 plates are only have chosen, and the independent variable of time series also accordingly sets
It is set to 9.Advantage of this is that, we are objective to consider every stock may have the possibility of risk exposure to multiple industries
Property so that our analyses of model are more comprehensive.Then model can be according to such stock for drawing of analyzing to the sudden and violent of plate
Dew, then gross section regression is done together with the factor exposure of other risks and assumptions, obtain being mapped in the exposure of the industry-by-industry factor
On regression coefficient, and be based ultimately upon these regression coefficients and calculate the plate factor current yield.
The present invention makes prediction according to the predictive risk following to investment combination of model, so as to help investor more preferable
Do good before risk management.It is different from the current technical problem for mainly focusing on risk management afterwards.In the present invention
Risk can not only be predicted and quantified to model according to different Factorizations, improve the stability of risk profile, and can be by
Investment risk decomposes separate sources, selects to undertake which risk so as to make investor more targeted and which wind evaded
Danger.
By the system, user only needs to the data input of the manual position in storehouse proportion by every stock of investment combination, for example
In investment combination, stock A holds ratio and accounts for investment combination 5%, and stock B accounts for 3% etc., you can carry out immediately investment combination and
The risk analysis and prediction of personal share.
The present invention makes prediction according to the predictive risk following to investment combination of model, so as to help investor more preferable
Do good before risk management.It is different from the current technical problem for mainly focusing on risk management afterwards.In the present invention
Risk can not only be predicted and quantified to model according to different Factorizations, improve the stability of risk profile, and can be by
Investment risk decomposes separate sources, selects to undertake which risk so as to make investor more targeted and which wind evaded
Danger.
By this method, user only needs to the data input of the manual position in storehouse proportion by every stock of investment combination, for example
In investment combination, stock A holds ratio and accounts for investment combination 5%, and stock B accounts for 3% etc., you can carry out immediately investment combination and
The risk analysis and prediction of personal share.
Embodiment 2
As a kind of preferred embodiment of the invention, present embodiment discloses a kind of movement in stock and share rate Forecasting Methodology, including
Following steps：
Step one, initial data is read from database, initial data includes that market data relevant with stock and finance are reported
Table data；
Step 2, set up stability bandwidth forecast model；
Step 3, the model calculating personal share volatitle revenue dynamic rate using foundation in step 2.
The computational methods of model are as follows in the step 2：
Assuming that the common k risks and assumptions of the multiple-factor risk model of earning rate.The then earning rate r of stock iiFor：
ri=bi1f1+bi2f2+bi3f3+....+bikfk+∈i (1)
Wherein bijIt is exposures of the stock i to risks and assumptions j；fjIt is the earning rate of risks and assumptions j, ∈iIt is the personal share of stock i
Income, i.e., expose that unrelated a part of income of brought income to risks and assumptions, different from the income of stock i with the stock
Rate ri；
Based on (1), the volatitle revenue dynamic rate of the stock is：
Wherein biIt is the k × 1 risks and assumptions exposure vector of stock i, V is k × k risks and assumptions yield variance-covariance square
Battle array, σεiThe stability bandwidth of personal share income, i.e. personal share risk.
Embodiment described above only expresses the specific embodiment of the application, and its description is more specific and detailed, but simultaneously
Therefore the limitation to the application protection domain can not be interpreted as.It should be pointed out that for one of ordinary skill in the art
For, on the premise of technical scheme design is not departed from, various modifications and improvements can be made, these belong to this
The protection domain of application.
Claims (9)
1. a kind of movement in stock and share rate Forecasting Methodology, it is characterised in that comprise the following steps：
Step one, from database read initial data, initial data include the market data and financial statement number relevant with stock
According to；
Step 2, set up stability bandwidth forecast model；
Step 3, the model calculating personal share volatitle revenue dynamic rate using foundation in step 2.
2. a kind of movement in stock and share rate Forecasting Methodology according to claim 1, it is characterised in that the computational methods of the model
It is as follows：
Assuming that the common k risks and assumptions of the multiple-factor risk model of earning rate.The then earning rate r of stock iiFor：
ri=bi1f1+bi2f2+bi3f3+....+bikfk+∈i (1)
Wherein bijIt is exposures of the stock i to risks and assumptions j；fjIt is the earning rate of risks and assumptions j, ∈iIt is the personal share income of stock i,
Risks and assumptions are exposed with that unrelated a part of income of brought income with the stock, different from the earning rate r of stock ii；
Based on (1), the volatitle revenue dynamic rate of the stock is：
Wherein biIt is the k × 1 risks and assumptions exposure vector of stock i, V is k × k risks and assumptions yield variance-covariance matrix, σεi
The stability bandwidth of personal share income, i.e. personal share risk.
3. a kind of movement in stock and share rate Forecasting Methodology according to claim 2, it is characterised in that：The exposure of the risks and assumptions
Computational methods are：After each transaction end of day initial data updates, every primitive factor exposure of stock is collected and extracts, it is false
If stock i exposes as B to factor k'sik, then standardization of the stock i to factor k is exposed as：
bik=(Bik-mean(Bk))/std(Bk) (3)
Wherein BikIt is stock i to the standardization of factor k exposure, B is exposed vector of all stocks to factor k in sample.
4. a kind of movement in stock and share rate Forecasting Methodology according to claim 2, it is characterised in that the income of the risks and assumptions
Rate calculation is：With the multiple power compound interest earning rate of every stock of 22 day of trade of past as dependent variable, with 22 day of trade
The preceding acquired factor is exposed as independent variable, then does linear regression based on these data.
5. a kind of movement in stock and share rate Forecasting Methodology according to claim 2, it is characterised in that the risks and assumptions income side
Difference-covariance matrix calculation procedure is：
After step one, the closing quotation of each day of trade, newest initial data is obtained；
Step 2, calculating factor exposure, then calculate same day factor earning rate；
Step 3, with the factor earning rate time series data in past 36 months come calculation risk factor variance-covariance matrix.
6. a kind of movement in stock and share rate Forecasting Methodology according to claim 1, it is characterised in that the stability bandwidth forecast model
Method for building up comprise the following steps：
Step one, from database read initial data, initial data include the market data and financial statement number relevant with stock
According to；
Step 2, according to initial data build risks and assumptions；
Step 3, the factor that will be built in step 2 are stored into factor data storehouse；
Step 4, with step 3 determine risks and assumptions set up stability bandwidth forecast model.
7. a kind of movement in stock and share rate Forecasting Methodology according to claim 6, it is characterised in that the risks and assumptions include：
Reverse the factor, factor of momentum, the market value factor, flowing sex factor, the debt ratio factor, the HSBC factor and the industry factor；It is described inverse
Transposon includes that stock goes over to weigh compound interest overall yield again in 1 month；It is multiple that the factor of momentum includes that stock goes over 12 months multiple power
Breath overall yield subtract over one month again power compound interest overall yield difference；The market value factor is included negotiable stock valency
Lattice are multiplied by the value obtained after the product extraction of square root of negotiable stock quantity；The flowing sex factor includes that stock went over for 30 day of trade
Transaction turnover rate, turnover rate be trading volume divided by negotiable total amount quotient；The debt ratio factor is looked forward to including previous year
Quotient of the industry total liability divided by previous year enterprise shareholding equity；The HSBC factor include previous year enterprise total market capitalisation divided by
The quotient of previous year enterprise shareholding equity；The industry factor includes the historical data based on the past 36 months of each plate to every
Stock does Multivariate Time Series recurrence, and the plate includes can active plate, material plate, industrial plate, durable goods plate
Block, optional commodity plate, medical treatment ＆ health plate, financial plate, information technology plate and public utilities plate.
8. a kind of movement in stock and share rate Forecasting Methodology according to claim 1 and 6, it is characterised in that：The market packet
Include the daily price data of stock and trading volume data.
9. a kind of movement in stock and share rate Forecasting Methodology according to claim 1 and 6, it is characterised in that：The financial statement number
According to each financial season including stock and the profit and loss statement data of wealth year, assets debt tables data and cash flow statement data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710173259.XA CN106934502A (en) | 2017-03-22 | 2017-03-22 | A kind of movement in stock and share rate Forecasting Methodology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710173259.XA CN106934502A (en) | 2017-03-22 | 2017-03-22 | A kind of movement in stock and share rate Forecasting Methodology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106934502A true CN106934502A (en) | 2017-07-07 |
Family
ID=59433539
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710173259.XA Pending CN106934502A (en) | 2017-03-22 | 2017-03-22 | A kind of movement in stock and share rate Forecasting Methodology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106934502A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019205381A1 (en) * | 2018-04-28 | 2019-10-31 | 平安科技（深圳）有限公司 | Stock screening method and device, and computer-readable storage medium |
-
2017
- 2017-03-22 CN CN201710173259.XA patent/CN106934502A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019205381A1 (en) * | 2018-04-28 | 2019-10-31 | 平安科技（深圳）有限公司 | Stock screening method and device, and computer-readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Giot et al. | Modelling daily value-at-risk using realized volatility and ARCH type models | |
| Gourio | Credit risk and disaster risk | |
| CN106934503A (en) | A kind of movement in stock and share rate forecasting system | |
| Jena et al. | The connectedness in the world petroleum futures markets using a Quantile VAR approach | |
| Stavroyiannis et al. | Econometric modeling and value-at-risk using the Pearson type-IV distribution | |
| Czudaj | Crude oil futures trading and uncertainty | |
| Wijaya | The effect of financial ratios toward stock returns among Indonesian manufacturing companies | |
| Huang | Takeover vulnerability and the credibility of signaling: The case of open-market share repurchases | |
| Engsted et al. | Return predictability and intertemporal asset allocation: Evidence from a bias-adjusted VAR model | |
| Palmborg et al. | Financial position and performance in IFRS 17 | |
| Cepni et al. | Local currency bond risk premia of emerging markets: The role of local and global factors | |
| CN106951996A (en) | A kind of method for building up of movement in stock and share rate forecast model | |
| CN106934502A (en) | A kind of movement in stock and share rate Forecasting Methodology | |
| Sakkas et al. | Forecasting the long-term equity premium for asset allocation | |
| Baaquie | A statistical model of the firm | |
| Sudhakar | Analysis of the efficacy of black-scholes model-an empirical evidence from call options on nifty-50 index | |
| Lian et al. | Portfolio selection in a multi-asset, incomplete-market economy | |
| Dwiyanthi et al. | The Influence of Financial Ratios on Share Prices (Case Studies on Property & Realestate Companies in the Construction and Building Sub-Sector Listed on the Indonesia Stock Exchange) | |
| Melan et al. | Ratio Analysis of Financials and Stock Price of Heavy Construction and Civil Engineering Companies | |
| Fu et al. | Hedging the portfolio of raw materials and the commodity under the mark-to-market risk | |
| Puchkova et al. | Fundamental analysis: the essence and necessity of application in portfolio investment (on the example of the Russian companies). | |
| Mustafa et al. | The Influence of Net Profit Margin, Dividend Payout Ratio, and Total Asset Turnover on Sharia Share Prices in Jakarta Islamic Index | |
| Ferrentino et al. | A Mathematical Model for the Pricing of Derivative Financial Products: the Role of the Banking Supervision and of the Model Risk | |
| SIVITSKA | Specificity of the discount rate application for agricultural business valuation. | |
| Hjartarson | Housing market dynamics: A non-rational approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170707 |